Vascular Medicine Stem Cell Mobilization Induced by Subcutaneous Granulocyte-Colony Stimulating Factor to Improve Cardiac Regeneration After Acute ST-Elevation Myocardial Infarction Result of the Double-Blind, Randomized, Placebo-Controlled Stem Cells in Myocardial Infarction (STEMMI) Trial Rasmus Sejersten Ripa, MD; Erik Jørgensen, MD; Yongzhong Wang, MD; Jens Jakob Thune, MD; Jens Christian Nilsson, MD; Lars Søndergaard, MD; Hans Erik Johnsen, MD; Lars Køber, MD; Peer Grande, MD; Jens Kastrup, MD Background—Phase 1 clinical trials of granulocyte-colony stimulating factor (G-CSF) treatment after myocardial infarction have indicated that G-CSF treatment is safe and may improve left ventricular function. This randomized, double-blind, placebo-controlled trial aimed to assess the efficacy of subcutaneous G-CSF injections on left ventricular function in patients with ST-elevation myocardial infarction. Methods and Results—Seventy-eight patients (62 men; average age, 56 years) with ST-elevation myocardial infarction were included after successful primary percutaneous coronary stent intervention ⬍12 hours after symptom onset. Patients were randomized to double-blind treatment with G-CSF (10 g/kg of body weight) or placebo for 6 days. The primary end point was change in systolic wall thickening from baseline to 6 months determined by cardiac magnetic resonance imaging (MRI). An independent core laboratory analyzed all MRI examinations. Systolic wall thickening improved 17% in the infarct area in the G-CSF group and 17% in the placebo group (P⫽1.0). Comparable results were found in infarct border and noninfarcted myocardium. Left ventricular ejection fraction improved similarly in the 2 groups measured by both MRI (8.5 versus 8.0; P⫽0.9) and echocardiography (5.7 versus 3.7; P⫽0.7). The risk of severe clinical adverse events was not increased by G-CSF. In addition, in-stent late lumen loss and target vessel revascularization rate in the follow-up period were similar in the 2 groups. Conclusions—Bone marrow stem cell mobilization with subcutaneous G-CSF is safe but did not lead to further improvement in ventricular function after acute myocardial infarction compared with the recovery observed in the placebo group. (Circulation. 2006;113:1983-1992.) Key Words: angiogenesis 䡲 heart failure 䡲 magnetic resonance imaging 䡲 myocardial infarction 䡲 stem cells
I
n patients with acute ST-elevation myocardial infarction (STEMI), treatment with acute percutaneous coronary stent intervention (PCI) is the method of choice to reestablish coronary blood flow. Although normal epicardial blood flow is reestablished within a few hours after symptom onset, myocardial damage is usually unavoidable and may result in heart failure caused by adverse left ventricular remodeling.1
myocardium by inducing myogenesis and vasculogenesis and improve left ventricular function.2 Most clinical studies have used intracoronary infusion of bone marrow mononuclear cells during cardiac catheterization after acute STEMI.3–5 However, the results are not conclusive, and the optimal route of stem cell administration remains to be determined. Prolonged pharmacological mobilization of bone marrow stem cells with granulocyte colony stimulating factor (GCSF) is an attractive alternative because the treatment is noninvasive and well known from clinical hematology.6 Phase 1 clinical trials after myocardial infarction have indi-
Editorial p 1926 Clinical Perspective p 1992 Animal studies indicate that treatment with bone marrow– derived adult stem cells after STEMI may regenerate
Received January 13, 2006; revision received February 23, 2006; accepted February 27, 2006. From the Department of Cardiology, The Heart Centre, University Hospital Rigshospitalet, Copenhagen (R.S.R., E.J., Y.W., J.J.T., L.S., L.K., P.G., J.K.); Department of Radiology, University Hospital Rigshospitalet, Copenhagen (R.S.R.); Danish Research Centre for Magnetic Resonance, University Hospital Hvidovre, Hvidovre (J.C.N.); and Department of Haematology, Aalborg University Hospital, Aalborg (H.E.J.), Denmark. The online-only Data Supplement can be found at http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.105.610469/DC1. The study has been registered in clinicaltrial.gov (NCT00135928). Correspondence to Jens Kastrup, MD, DMSc, Medical Department B, Cardiac Catheterization Laboratory 2014, The Heart Centre, University Hospital Rigshospitalet, DK-2100 Copenhagen Ø, Denmark. E-mail:
[email protected] © 2006 American Heart Association, Inc. Circulation is available at http://www.circulationaha.org
DOI: 10.1161/CIRCULATIONAHA.105.610469
1983
1984
Circulation
April 25, 2006
Figure 1. STEMMI trial flowchart.
cated that G-CSF treatment is safe and may improve left ventricular function.7–10 However, none of these were doubleblinded and placebo-controlled trials; thus, the effect of G-CSF–induced mobilization of bone marrow– derived stem cells on left ventricular function remains unknown.11 The aim of the present Stem Cells in Myocardial Infarction (STEMMI) trial was to asses, in a prospective, double-blind, randomized, placebo-controlled study, the efficacy of subcutaneous G-CSF injections on the regional and global left ventricular myocardial function in patients with STEMI and successful primary PCI.
The treatment was initiated from 1 to 2 days after the STEMI, and 78 patients were included from June 2003 to January 2005, as illustrated in Figure 1. Four patients withdrew consent before completion of G-CSF/placebo treatment as a result of severe claustrophobia during the baseline magnetic resonance imaging (MRI). MRI was not feasible in another 16 patients because of claustrophobia or obesity; these patients remained in the trial and were followed up per protocol with echocardiography and clinical and invasive examinations. Three patients refused follow-up examinations. The authors had full access to the data and take full responsibility for its integrity. All authors have read and agree to the manuscript as written.
End Points
Methods Study Population Seventy-eight patients with STEMI (62 men, 16 women; age, 56⫾8 years, mean⫾SD) who had been treated successfully with primary PCI within 12 hours after the onset of symptoms were included in the study. STEMI was diagnosed from typical chest pain at rest lasting ⬎30 minutes, the presence of cumulative ST-elevations ⬎0.4 mV in ⱖ2 contiguous leads on a standard 12-lead ECG, and a significant rise in serum markers of myocardial infarction. Only patients who were between 20 and 70 years of age with a culprit lesion located in the proximal section of a large coronary artery branch, plasma creatine kinase-MB ⬎100 g/L, or development of significant Q waves in the ECG were included. Exclusion criteria were prior myocardial infarction, significant stenosis in a nonculprit coronary vessel, ventricular arrhythmia after PCI requiring treatment, pregnancy, unprotected left main stem lesion, diagnosed or suspected cancer, New York Heart Association class 3 to 4 heart failure symptoms, or known severe claustrophobia. The study was approved by the local ethical committee (KF 01–239/02) and the Danish Medicines Agency (2612–2225). All patients received oral and written information about the study and signed an informed consent before inclusion in the study.
Study Design The STEMMI trial was a prospective, double-blind, randomized, placebo-controlled study allocating patients with acute STEMI in a 1:1 ratio after the primary PCI to G-CSF or placebo as a supplement to treatment according to guidelines. Randomization was done in blocks of 4 patients through the use of sequentially numbered, sealed envelopes. Patients received double-blinded treatment with either G-CSF (Neupogen, Amgen Europe BV, Breda, The Netherlands; 10 g/kg body weight) or a similar volume of placebo (isotonic sodium-chloride) as a subcutaneous injection once daily for 6 days.
The prespecified primary end point was change in regional systolic wall thickening from day 1 to 6 months evaluated with cardiac MRI. Secondary end points were (1) change in ejection fraction, end-systolic and end-diastolic volumes, and infarct size by MRI and (2) change in ejection fraction and end-systolic and end-diastolic volumes by echocardiography. Safety end points were (1) death of any cause, reinfarction, and new revascularization; (2) other adverse events; (3) in-stent restenosis; and (4) changes in inflammatory parameters (C-reactive protein and erythrocyte sedimentation rate). The 30-day clinical safety data have been published previously.12
Cardiac MRI MRI was performed before and 1 and 6 months after inclusion. A detailed description of the method is available in the online supplement 1. In short, cine images and late contrast-enhanced images were obtained with a 1.5-T clinical scanner (Siemens Vision Magnetom, Siemens AG, Erlangen, Germany). The independent core laboratory (Bio-Imaging Technologies BV, Leiden, the Netherlands) analyzed all examinations using the MRIMASS version 6.1 (MEDIS Medical Imaging Systems, Leiden, the Netherlands). The core laboratory was blinded to all patient data and to the order of the follow-up examinations. The analyses of cardiac MRI have a low interobserver variability (3% to 6%); to reduce this variability even more, each analysis was quality controlled by a second MRI technician before final approval. Regional left ventricular function was assessed by systolic wall thickening in the infarct region, the border region, and normal myocardium. Relative systolic wall thickening was calculated as the difference between diastolic and systolic divided by the diastolic wall thickness. Left ventricular end-diastolic volumes, end-systolic volumes, and myocardial mass were automatically calculated by the software. Infarct mass was quantified by selecting the signal intensity threshold of the hyperenhanced area.
Ripa et al TABLE 1.
Stem Cells in Myocardial Infarction
1985
Baseline Characteristics Placebo (n⫽39)
G-CSF (n⫽39)
Age, y
54.7⫾8.1
57.4⫾8.6
Male gender, n (%)
34 (87)
28 (72)
Body mass index, kg/m
28.1⫾3.9
27.4⫾4.4
Current smoker, n (%)
31 (80)
22 (56)
2
Diabetes mellitus, n (%) Known hypertension, n (%) Family history of ischemic heart disease, n (%)
4 (10)
3 (8)
10 (26)
13 (33)
14 (37)
15 (40)
4:23 (3:32–6:44)
3:47 (2:43–5:25)
LAD
13 (33)
21 (54)
LCx
7 (18)
5 (13)
RCA
19 (49)
13 (33)
TIMI flow grade 0 or 1 before primary PCI
34 (87)
33 (85)
TIMI flow grade 3 after primary PCI
39 (100)
36 (92)
2 (5)
1 (3)
Bare metal
26 (67)
25 (64)
Drug eluting
11 (28)
13 (33)
18 (46)
14 (36)
4 (10)
9 (23)
274 (141–441)
320 (194–412)
27:50 (22:34–36:16)
29:37 (25:21–44:15)
38 (100)
39 (100)
Median time from symptom debut to primary PCI (quartiles), h Index infarct-related artery, n (%)
Type of stent, n (%) No stent
Platelet glycoprotein IIb/IIIa inhibitors, n (%) PCI of second vessel during acute procedure, n (%) Median maximum serum CK-MB concentration (quartiles),* g/L Median time from PCI to G-CSF/placebo (quartiles), h Discharge medication,† n (%) Aspirin Clopidogrel
37 (97)
39 (100)
-Blocker
33 (87)
33 (85)
Statin
38 (100)
39 (100)
ACE inhibitor
15 (40)
20 (51)
Aspirin
33 (97)
37 (100)
Clopidogrel
33 (97)
37 (100)
-Blocker
27 (79)
31 (84)
Statin
34 (100)
37 (97)
ACE inhibitor
21 (62)
24 (65)
Medication at 6-mo‡ follow-up
LAD indicates left anterior descending artery; LCx, left circumflex artery; RCA, right coronary artery; and CK-MB, creatine kinase myocardial band. Data are mean⫾SD when appropriate. *Normal upper limit⫽5. †One in-hospital death in the placebo group. ‡Five and 2 patients lost to follow-up, respectively.
Echocardiography
Quantitative Coronary Angiography
Two-dimensional echocardiography was performed in 55 patients at baseline and after 6 months of follow-up. All patients were examined in the left recumbent position with a Vivid7 scanner (GE Medical Systems, Horten, Norway). Left ventricular volumes at end systole and end diastole were assessed by Simpson’s biplane method. Left ventricular ejection fraction was assessed in multiples of 5% by visual assessment by 1 experienced echocardiogram reader blinded to all patient data.
All patients were scheduled to undergo coronary angiography and, if required, repeated PCI before the 6-month MRI follow-up. All coronary angiograms for quantitative coronary angiography (QCA) analysis were acquired after intracoronary injection of 0.2 mg glyceryl nitrate according to guidelines. The independent core laboratory (Bio-Imaging Technologies BV) performed the QCA analysis using the QCA-CMS version 5.3 software (MEDIS Medical Imaging Systems). Before analysis, the core laboratory conducted
1986
Circulation
April 25, 2006
Figure 2. Changes during the first month after STEMI of (A) white blood cell count and concentration of CD34⫹ cells and plasma concentration of (B) SDF-1 and (C) VEGF-A. Bars indicate mean⫾SE.
the standardized frame selection on the angiograms according to the worst view. The selected frames were end diastolic, showed minimal foreshortening, had no overlap, and had good contrast. All frame selections and QCA analyses were conducted according to established standard operating procedures. After each analysis, a second QCA technician performed the quality control before final approval. The QCA technicians were fully blinded to all patient data.
Analyses of Peripheral Blood Samples of venous blood were obtained before the G-CSF/placebo treatment (baseline), at days 4 and 7, and at 1 month. The concentrations of CD34⫹ cells, CD45⫺/CD34⫺, and subpopulations in the peripheral blood were measured by flow cytometry as previously described.13 Plasma concentrations of vascular endothelial growth factor A (VEGF-A) and stromal cell-derived factor 1 (SDF-1) concentrations were measured in duplicate by a colorimetric ELISA kit (R&D Systems, Minneapolis, Minn). The lower limits of detection were 10 pg/mL for VEGF-A and 18 pg/mL for SDF-1.
Statistical Analysis The pretrial power calculation showed that a sample size of 50 would yield an expected power of ⬎90% to detect a difference of 15 percentage points between the treated and the placebo groups, with a 2-sided significance level of 0.05, and an assumed standard deviation of 15 percentage points for the systolic wall thickening change from baseline to 6 months in both groups. To allow for 33% dropout and claustrophobia or difficulties with breathholding during the very early baseline MRI, we decided on a sample size of 78 patients. Analyzes were performed on an intention to treat basis. The effects of the G-CSF treatment were analyzed in a 2-factor ANOVA with repeated measures as a within-subject factor and treatment group as a between-subjects factor or Student’s t test comparing changes from baseline to 6 months follow-up in the G-CSF and placebo groups. Subgroup analyses were not prespecified but based on recent results from the Reinfusion of Enriched Progenitor cells And Infarct
Ripa et al TABLE 2.
Stem Cells in Myocardial Infarction
1987
G-CSF–Induced Cell Mobilization Placebo (n⫽33)
G-CSF (n⫽37)
P
CD45⫺/CD34⫺, per 1 L Baseline
246⫾164
253⫾205
0.7
Day 7
260⫾202
1015⫾1114
⬍0.001
Day 28
183⫾134
169⫾137
0.5
Baseline
14⫾12
22⫾48
0.6
Day 7
15⫾15
42⫾45
⬍0.001
Day 28
6⫾7
9⫾12
0.5
CD45⫺/CD34⫺/CXCR4⫹, per 1 L
CD45⫺/CD34⫺/VEGFR-2⫹, per 1 L Baseline
4⫾3
5⫾6
1.0
Day 7
3⫾3
30⫾47
⬍0.001
Day 28
2⫾2
3⫾3
0.8
Data are mean⫾SD.
Remodeling in Acute Myocardial Infarction (REPAIR-AMI) trial.3 Analyses were performed with SPSS (version 12.0, SPSS Inc, Chicago, Ill). The level of statistical significance was set at P⬍0.05, except for the subgroup analyses (see online-only Data Supplement Table), for which significance was set at P⬍0.01 to account for multiple testing.
Results The clinical and angiographic characteristics of the study population are shown in Table 1. The groups were homogeneous with a trend toward more smokers and slightly longer time from symptom onset to primary PCI in the placebo group. All patients received optimal medical treatment in the follow-up period (Table 1). Subcutaneous G-CSF or placebo was initiated 10 to 65 hours (with 85% initiated ⬍48 hours) after the primary PCI, resulting in a rapid parallel mobilization of leukocytes and CD34⫹ cells into the peripheral circulation in the G-CSF group (Figure 2A), whereas the cells remained unchanged or decreased slightly in the placebo group. In addition, treatment with G-CSF resulted in mobilization of a mesenchymal stem cell line CD45⫺/CD34⫺ cells expressing the SDF-1 receptor CXCR-4 and the VEGF receptor 2 (Table 2). The plasma concentration of SDF-1 increased rapidly and significantly within the first days after the STEMI in the placebo group and remained elevated in the first month (Figure 2B). In contrast, SDF-1 was unchanged during G-CSF treatment and increased significantly at the 1-month follow-up. VEGF-A fluctuated nonsignificantly without differences between the placebo and G-CSF groups (Figure 2C).
Treatment Effect of G-CSF on Left Ventricular Function Change in systolic wall thickening in the infarct area from baseline to the 6-month follow-up did not differ significantly between the placebo and G-CSF groups (17⫾32 versus 17⫾22 percentage points; Figure 3A and 3D). The corresponding changes in systolic wall thickening in the infarct border zone (Figure 3B and 3D) and in noninfarcted normal myocardium (Figure 3C and 3D) tended to be lower in the G-CSF group (8⫾23 and ⫺2⫾30 percentage points) com-
pared with the placebo group (23⫾23 and 19⫾38 percentage points). However, these differences can be attributed to differences in the baseline values because the values were similar in the 2 groups at the 6-month follow-up. No significant differences were identified between the 2 groups in terms of end-diastolic volume, end-systolic volume, left ventricular mass, and left ventricular ejection fraction derived from the MRI examinations (Figure 4). These results are further confirmed by similar changes in the placebo and G-CSF groups measured with echocardiography (Figure 5). In addition, baseline measures of global left ventricular morphology and function were very similar in the 2 groups (Table 3). The homogeneity of left ventricular ejection fraction changes measured with MRI between the placebo and G-CSF groups was consistent in a number of subgroups analyzed (Data Supplement Table). The initial infarct size measured by late contrast enhancement MRI was similar in the G-CSF (median, 8 g; range, 1 to 31 g) and placebo (median, 9 g; range, 1 to 37 g) groups. The infarct sizes were unchanged in the 2 groups from baseline to the 6-month follow-up (Figure 6A). Similar results were found for infarct size in percent of left ventricular mass (Figure 6B).
Safety of Treatment With G-CSF Soon After STEMI G-CSF treatment was well tolerated, with a trend toward more patients in the G-CSF group reporting mild to moderate musculoskeletal pain (Table 4) during the treatment. No patients withdrew consent because of this side effect. C-reactive protein and erythrocyte sedimentation rates were analyzed as markers of inflammation. C-reactive protein was elevated similarly at baseline in the 2 groups (median, 17 and 19 mg/L) and subsequently normalized in the placebo group, whereas the concentration increased slightly during G-CSF treatment (median at day 4, 34 mg/L; at day 7, 22 mg/L) but had normalized at the 1-month follow-up (G-CSF treatment effect, P⫽0.07). Blood sedimentation rates showed a similar pattern in the G-CSF and placebo groups (P⫽0.2).
1988
Circulation
April 25, 2006
Figure 4. Change in global left ventricular volumes and mass from baseline to 6-month follow-up measured with MRI. Bars indicate mean⫾SE.
Figure 3. Regional systolic wall thickening measured with MRI at baseline and 1- and 6-month follow-up. Bars indicate mean⫾SE.
As previously reported,12 there were 2 in-hospital major adverse cardiac events (Table 4). One patient in the placebo group progressed into cardiogenic shock after the primary PCI and died 2.5 days later, despite aggressive treatment (intra-aortic balloon pump, dialysis, and ventilator therapy). At the time of death, the patient had received a total of 2 injections of placebo. One patient in the G-CSF group had subacute stent thrombosis 2 days after the primary PCI. The patient initially had a thrombotic total occlusion of the distal right coronary artery, which was treated with a bare metal stent (18⫻4 mm), resulting in TIMI grade 3 flow. Forty-eight hours later, the patient had recurrent chest pain and reelevation of the ST-segment on the ECG but no recurrent increase in biochemical markers. An acute angiogram showed thrombotic occlusion at the proximal edge of the stent. This occlusion was treated with a new stent implantation. TIMI grade 3 flow was restored, and there was complete ST resolution. The subacute stent thrombosis occurred ⬇6 hours after the first subcutaneous injection of G-CSF. No sustained ventricular arrhythmias were detected during in-hospital telemetric monitoring. No additional death, reinfarction, or stent thrombosis occurred in the follow-up period (Table 4). However, 2 patients in the placebo group were referred for heart surgery. One patient had mitral valve repair, and 1 patient underwent coronary artery bypass surgery as a result of progressive symptoms of coronary artery disease. Coronary angiograms were obtained in 31 placebo (80%) and 35 G-CSF (90%) patients on average 5 months after the initial PCI. Target vessel revascularization was performed in 8 patients (12%) in relation to the angiographic follow-up (4 patients in the placebo group, 4 in the G-CSF group; P⫽1.0), whereas non–target vessel revascularization was performed in 4 (13%) patients in the placebo group and 2 (6%) in the G-CSF group (P⫽0.4) (Table 4). The results of the quantitative coronary angiography are shown in Table 5; 5 angiograms were not analyzable for technical reasons. There were no differences between groups at baseline and at follow-up.
Ripa et al
Stem Cells in Myocardial Infarction
1989
Figure 5. Change in global left ventricular volumes from baseline to 6-month follow-up measured with echocardiography. Bars indicate mean⫾SE.
Discussion This first double-blind, randomized, placebo-controlled trial demonstrated that subcutaneous G-CSF mobilization of bone marrow stem cells had no effect on left ventricular myocardial function or infarct size after STEMI treated with primary PCI. However, treatment with G-CSF seemed safe and well tolerated. G-CSF is a potent hematopoietic cytokine that increases the production of granulocytes and is involved in mobilization of granulocytes and stem and progenitor cells from the bone marrow into the blood circulation.14 The mobilization process is not fully understood but is mediated through enzyme release, leading to digestion of adhesion molecules, and through trophic chemokines; SDF-1 and its receptor CXCR-4 seem of paramount importance.15 Animal studies and phase 1 clinical trials have suggested a beneficial effect of G-CSF on left ventricular function after myocardial infarction. Orlic et al2 reported favorable results after stem cell mobilization with G-CSF and stem cell factor in mice with acute myocardial infarction. A recent experiment with G-CSF after reperfused myocardial infarction in rabbits showed an improvement in left ventricular ejection fraction and reduced remodeling.16 Kuethe et al8 compared 14 patients treated with G-CSF 2 days after STEMI with 9 patients who refused G-CSF treatment. The treated group had a nonsignificantly higher increase in ejection fraction compared with the control group.
Similar results were found in a single-blinded, placebocontrolled study including 20 patients 1.5 days after STEMI.7 The Front-Integrated Revascularization and Stem Cell Liberation in Evolving Acute Myocardial Infarction by Granulocyte Colony Stimulating Factor (FIRSTLINE-AMI) trial was a phase 1 randomized, open-label trial of 50 patients. G-CSF treatment was initiated 1.5 hours after STEMI. The control group did not receive placebo. The trial suggested improvement in left ventricular function with enhanced wall thickening, improvement in ejection fraction, and no change in end-diastolic diameter. The control group had less systolic wall thickening, decreased ejection fraction, and increased end-diastolic diameter.9 Thus, several studies support the efficacy of G-CSF treatment after STEMI, suggesting an
TABLE 3. Baseline Left Ventricular Morphology and Function Measured by MRI and Echocardiography
MRI, n Ejection fraction,% End-diastolic volume, mL
Placebo
G-CSF
29
28
P
55.7⫾9.8
51.2⫾15.4
0.2
127.4⫾28.3
125.5⫾36.5
0.8
End-systolic volume, mL
57.5⫾22.5
64.0⫾36.5
0.4
Left ventricular mass, g
186.2⫾39.0
181.8⫾46.8
0.7
Echocardiography, n
26
29
48.9⫾10.6
49.9⫾10.8
0.7
End-diastolic volume, mL
100.1⫾28.8
90.3⫾26.8
0.2
End-systolic volume, mL
49.7⫾17.9
45.4⫾18.9
0.4
Ejection fraction,%
Data are mean⫾SD.
Figure 6. Infarct size in (A) grams and (B) percent of left ventricular mass measured with MRI. Bars indicate mean⫾SE. LV indicates left ventricular; AMI, acute myocardial infarction.
1990
Circulation
April 25, 2006
TABLE 4. Cumulative Clinical Events in the 6-Month Follow-Up Period Placebo (n⫽39), n
G-CSF (n⫽39), n
1
0
Possible explanations for the absence of an additional improvement in left ventricular function, despite a very significant G-CSF mobilization of CD34⫹ and CD45⫺/ CD34⫺ mononuclear cells with the potential for homing to the necrotic areas, are lack of homing signals from the myocardium, too low a dose of G-CSF, wrong timing of the treatment, mobilization of inactive subsets of stem cell populations, and use of inappropriate end points. Plasma concentration of stromal cell– derived factor-1 (SDF-1) is known to increase rapidly during the first week after a myocardial infarction and to reach a maximum concentration after 3 weeks,18 which is in concordance with our findings in the placebo group. The SDF-1 expression is thought to play a crucial role in induction of stem cell engraftment to ischemic tissue.19 However, we found an unchanged concentration of SDF-1 during the G-CSF treatment. This could be due to an inhibition of production of homing signals from the myocardium or to the consumption of SDF-1 by cells that engrafted to the infarcted myocardium. In addition, the quantity of membrane-bound SDF-1, which may be the key mediator of homing, was unknown. Our trial was not designed to investigate either the optimal dose or the optimal time point for G-CSF treatment. The treatment regimen resulted in a maximum concentration of CD34⫹ mononuclear cells 4 to 7 days after the STEMI, which corresponds well to the recent results from the REPAIR-AMI trial that indicated that the optimal time for intracoronary infusion of bone marrow mononuclear cells is day 5 to 6 after the infarction.3 However, considering that plasma SDF-1 and VEGF-A reach a maximum 3 weeks after the infarction, one can speculate whether G-CSF treatment should have been postponed until this period. In contrast, a recent study in mice has indicated that G-CSF inhibits apoptosis of cardiomyocytes by directly affecting the cells rather than through mobilization of bone marrow cells.20 The study further indicated that the antiapoptotic effect of G-CSF was significantly reduced if treatment was delayed to 3 days after the infarction.20 This concept is interesting because the FIRSTLINE-AMI treated patients 1.5 hours after the STEMI and found an improvement in left ventricular function.9,10 We can only speculate if the difference in time to G-CSF can account for some of the difference in outcome.
In-hospital events Death Reinfarction
0
0
Stent thrombosis
0
1
Documented ventricular arrhythmia
0
0
Musculoskeletal discomfort*
3
8
Death
1
0
Reinfarction
0
0
Stent thrombosis
0
1
Thoracic surgery
2
0
Target vessel revascularization
4
4
Non–target vessel revascularization
4
2
Documented ventricular arrhythmia
0
0
6-Month follow-up (cumulative)
Cumulative death or reinfarction
1
0
Cumulative death, reinfarction, or target vessel revascularization
5
4
*P⫽0.2
improvement in ejection fraction of 6% to 8%, improved systolic wall thickening in the infarct zone, and unchanged end-diastolic volume. When our results, analyzed by a blinded, independent core laboratory, are reviewed, it is remarkable that the changes in our G-CSF group are comparable to the results of previous nonblinded phase 1 studies, whereas there are major differences in the control/placebo groups. The changes in global left ventricular function in the placebo group we report are comparable to those of the MRI study by Baks et al,17 who found an increase in ejection fraction of 7%, an unchanged end-systolic volume, and a small increase in end-diastolic volume following primary stent PCI after STEMI. Thus, an important finding of our trial is the prominent effect of primary stent PCI on left ventricular function in the treatment of STEMI. These findings emphasize the need for caution in the interpretation of positive results of phase 1 G-CSF trials. TABLE 5.
Angiographic Core Laboratory Results In-Lesion Zone
In-Stent Zone
G-CSF (n⫽32)
Placebo (n⫽30)
P
G-CSF (n⫽32)
Placebo (n⫽30)
P
3.04⫾0.60
2.99⫾0.55
0.7
3.06⫾0.55
2.98⫾0.56
0.6
After primary procedure
2.66⫾0.57
2.68⫾0.54
0.9
2.78⫾0.49
2.71⫾0.52
0.6
Follow-up
2.14⫾0.72
2.00⫾0.71
0.5
2.26⫾0.71
2.10⫾0.74
0.4
After primary procedure
12.4⫾14.1
11.1⫾9.9
0.7
8.4⫾11.5
9.0⫾7.9
0.8
Follow-up
28.0⫾22.2
31.4⫾19.0
0.5
23.42⫾22.6
28.1⫾20.0
0.4
0.43⫾0.59
0.69⫾0.64
0.1
0.45⫾0.50
0.63⫾0.64
0.3
3 (10)
4 (13)
0.7
3 (10)
3 (10)
1.0
Reference vessel diameter, mm Minimal lumen diameter, mm
Diameter stenosis, %
Late lumen loss, mm Binary (⬎50%) restenosis, n (%)
Data are mean⫾SD when appropriate.
Ripa et al We studied a homogeneous patient population without numerous factors that would tend to confound the results, but this has probably also led to exclusion of high-risk patients who would potentially benefit most from the treatment. Only patients with a significant rise in biochemical markers or an ECG indicating a large myocardial infarction were included. In addition, we had an upper age limit of 70 years because several previous trials have indicated that the mobilization of CD34⫹ cells decreases with increasing age.21 To further increase the power of our study, we chose regional myocardial function rather than global myocardial function as a primary end point. Thus, we consider the present findings very “robust” despite the discrepancy in results when compared with previous uncontrolled or unblinded phase 1 G-CSF trials, but we cannot exclude the possibility that the trial has been underpowered to detect a very small difference. The complex interaction between stem cell mobilization and cytokines remains poorly understood, and the results do not exclude the possibility that G-CSF could be part of a treatment strategy combing several cytokines and/or local stem cell delivery in future trials.
Safety The trial indicates good short-term safety with G-CSF treatment. The subcutaneous injections were well tolerated, with few patients experiencing mild musculoskeletal pain. There were no indications of excessively increased progression of atherosclerosis in the G-CSF group by angiography and no deaths or myocardial infarctions in patients treated with G-CSF. We found no indications of clinical significant change in blood viscosity caused by the increase in white blood cell count. This is in concordance with the unaltered viscosity measured in the FIRSTLINE-AMI trial.9 Noteworthy, as in recent trials7,9 we found no evidence of clinically important worsening of myocardial inflammation despite increases in inflammatory markers in patients treated with G-CSF. This is in agreement with our findings for G-CSF treatment in patients with chronic myocardial ischemia.22 We cannot exclude the possibility that G-CSF may have contributed to the 1 case of subacute stent thrombosis observed; however, we find this possibility very unlikely because of the short time period between the first subcutaneous injection of G-CSF and the occurrence of the stent thrombosis. Target vessel revascularization as a result of restenosis and late lumen loss in the G-CSF and placebo groups was low and within the expected range, considering the low proportion of diabetics in the patient cohort, typical for a Scandinavian cohort.23,24 Thus, the present coronary angiographic data analyzed by an independent core laboratory, combined with similar recent trials,7–9,25 argue strongly against speculations that G-CSF might enhance the restenosis process, as previously suggested.26 Further examination of safety should include treatment of more patients and a longer follow-up. In conclusion, subcutaneous injections of G-CSF soon after STEMI treated with primary PCI was well tolerated and seemed safe. However, there was no additional G-CSF treatment effect on the prespecified primary end points, addressing regional myocardial function, compared with the substantial recovery found in the placebo group. In addition,
Stem Cells in Myocardial Infarction
1991
secondary end points of global left ventricular function and infarct size were unaffected by the G-CSF treatment. The discrepancy in results between previous open-label and this double-blind, placebo-controlled study seems to underscore the need for blinding and for placebo controls in the evaluation of new, potentially angiogenic, stem cell therapies.
Acknowledgments The STEMMI trial has been funded with grants from the Danish Heart Foundation (No. 0442B18-A1322141); Danish Stem Cell Research Doctoral School; Faculty of Health Science, Copenhagen University; Research Foundation of Rigshospitalet; Raimond og Dagmar Ringgård-Bohn’s Foundation; Aase og Ejnar Danielsens Foundation; Erik og Martha Scheibels Legat; and Arvid Nilssons Foundation.
Disclosures None.
References 1. Pfeffer MA, Braunwald E. Ventricular remodeling after myocardial infarction: experimental observations and clinical implications. Circulation. 1990;81:1161–1172. 2. Orlic D, Kajstura J, Chimenti S, Limana F, Jakoniuk I, Quaini F, NadalGinard B, Bodine DM, Leri A, Anversa P. Mobilized bone marrow cells repair the infarcted heart, improving function and survival. Proc Natl Acad Sci U S A. 2001;98:10344 –10349. 3. Schächinger V, Erbs S, Elsässer A, Haberbosch W, Hambrecht R, Hölschermann H, Yu J, Corti R, Mathey D, Hamm C, Süselbeck T, Assmus B, Tonn T, Dimmeler S, Zeiher AM. Intracoronary infusion of bone marrow-derived progenitor cells in acute myocardial infarction: a randomized, double-blind, placebo-controlled multicenter trial (REPAIR-AMI). Circulation. 2005;112:3362. Abstract. 4. Strauer BE, Brehm M, Zeus T, Kostering M, Hernandez A, Sorg RV, Kogler G, Wernet P. Repair of infarcted myocardium by autologous intracoronary mononuclear bone marrow cell transplantation in humans. Circulation. 2002;106:1913–1918. 5. Wollert KC, Meyer GP, Lotz J, Ringes-Lichtenberg S, Lippolt P, Breidenbach C, Fichtner S, Korte T, Hornig B, Messinger D, Arseniev L, Hertenstein B, Ganser A, Drexler H. Intracoronary autologous bonemarrow cell transfer after myocardial infarction: the BOOST randomised controlled clinical trial. Lancet. 2004;364:141–148. 6. Johnsen HE. Clinical practice and future needs in recombinant human granulocyte colony-stimulating factor treatment: a review of randomized trials in clinical haemato-oncology. J Int Med Res. 2001;29:87–99. 7. Valgimigli M, Rigolin GM, Cittanti C, Malagutti P, Curello S, Percoco G, Bugli AM, Della PM, Bragotti LZ, Ansani L, Mauro E, Lanfranchi A, Giganti M, Feggi L, Castoldi G, Ferrari R. Use of granulocyte-colony stimulating factor during acute myocardial infarction to enhance bone marrow stem cell mobilization in humans: clinical and angiographic safety profile. Eur Heart J. 2005;26:1838 –1845. 8. Kuethe F, Figulla HR, Herzau M, Voth M, Fritzenwanger M, Opfermann T, Pachmann K, Krack A, Sayer HG, Gottschild D, Werner GS. Treatment with granulocyte colony-stimulating factor for mobilization of bone marrow cells in patients with acute myocardial infarction. Am Heart J. 2005;150:115. 9. Ince H, Petzsch M, Kleine HD, Schmidt H, Rehders T, Korber T, Schumichen C, Freund M, Nienaber CA. Preservation from left ventricular remodeling by front-integrated revascularization and stem cell liberation in evolving acute myocardial infarction by use of granulocytecolony-stimulating factor (FIRSTLINE-AMI). Circulation. 2005;112: 3097–3106. 10. Ince H, Petzsch M, Kleine HD, Eckard H, Rehders T, Burska D, Kische S, Freund M, Nienaber CA. Prevention of left ventricular remodeling with granulocyte colony-stimulating factor after acute myocardial infarction: final 1-year results of the Front-Integrated Revascularization and Stem Cell Liberation in Evolving Acute Myocardial Infarction by Granulocyte Colony-Stimulating Factor (FIRSTLINE-AMI) Trial. Circulation. 2005; 112(suppl I):I-73–I-80.
1992
Circulation
April 25, 2006
11. de Muinck ED, Simons M. Calling on reserves: granulocyte colony stimulating growth factor in cardiac repair. Circulation. 2005;112: 3033–3035. 12. Ripa RS, Wang Y, Jørgensen E, Johnsen HE, Grande P, Kastrup J. Safety of bone-marrow stem cell mobilisation induced by granulocyte-colony stimulating factor: clinical 30 days blinded results from the Stem Cells in Myocardial Infarction (STEMMI) trial. Heart Drug. 2005;5:177–182. 13. Mancuso P, Burlini A, Pruneri G, Goldhirsch A, Martinelli G, Bertolini F. Resting and activated endothelial cells are increased in the peripheral blood of cancer patients. Blood. 2001;97:3658 –3661. 14. Anderlini P, Donato M, Chan KW, Huh YO, Gee AP, Lauppe MJ, Champlin RE, Korbling M. Allogeneic blood progenitor cell collection in normal donors after mobilization with filgrastim: the M.D. Anderson Cancer Center experience. Transfusion. 1999;39:555–560. 15. Flomenberg N, DiPersio J, Calandra G. Role of CXCR4 chemokine receptor blockade using AMD3100 for mobilization of autologous hematopoietic progenitor cells. Acta Haematol. 2005;114:198 –205. 16. Minatoguchi S, Takemura G, Chen XH, Wang N, Uno Y, Koda M, Arai M, Misao Y, Lu C, Suzuki K, Goto K, Komada A, Takahashi T, Kosai K, Fujiwara T, Fujiwara H. Acceleration of the healing process and myocardial regeneration may be important as a mechanism of improvement of cardiac function and remodeling by postinfarction granulocyte colonystimulating factor treatment. Circulation. 2004;109:2572–2580. 17. Baks T, van Geuns RJ, Biagini E, Wielopolski P, Mollet NR, Cademartiri F, Boersma E, van der Giessen WJ, Krestin GP, Duncker DJ, Serruys PW, de Feyter PJ. Recovery of left ventricular function after primary angioplasty for acute myocardial infarction. Eur Heart J. 2005;26:1070 –1077. 18. Wang Y, Johnsen HE, Mortensen S, Bindslev L, Ripa RS, HaackSorensen M, Jørgensen E, Fang W, Kastrup J. Changes in circulating mesenchymal stem cells, stem cell homing factor, and vascular growth factors in patients with acute ST-elevation myocardial infarction treated with primary percutaneous coronary intervention. Heart. October 26, 2005. DOI: 10.1136/hrt.2005.069799. Available at: http://heart.bmjjournals.com/. Accessed 2006. 19. Askari AT, Unzek S, Popovic ZB, Goldman CK, Forudi F, Kiedrowski M, Rovner A, Ellis SG, Thomas JD, DiCorleto PE, Topol EJ, Penn MS. Effect of stromal-cell-derived factor 1 on stem-cell homing and tissue regeneration in ischaemic cardiomyopathy. Lancet. 2003;362:697–703. 20. Harada M, Qin Y, Takano H, Minamino T, Zou Y, Toko H, Ohtsuka M, Matsuura K, Sano M, Nishi J, Iwanaga K, Akazawa H, Kunieda T, Zhu
21.
22.
23.
24.
25.
26.
W, Hasegawa H, Kunisada K, Nagai T, Nakaya H, Yamauchi-Takihara K, Komuro I. G-CSF prevents cardiac remodeling after myocardial infarction by activating the Jak-Stat pathway in cardiomyocytes. Nat Med. 2005;11:305–311. de la Rubia J, Arbona C., de Arriba F, del Canizo C, Brunet S, Zamora C, Diaz MA, Bargay J, Petit J, de la SJ, Insunza A, Arrieta R, Pascual MJ, Serrano D, Sanjuan I, Espigado I, Alegre A, Martinez D, Verdeguer A, Martinez C, Benlloch L, Sanz MA. Analysis of factors associated with low peripheral blood progenitor cell collection in normal donors. Transfusion. 2002;42:4 –9. Wang Y, Tägil K, Ripa RS, Nilsson JC, Carstensen S, Jørgensen E, Sondergaard L, Hesse B, Johnsen HE, Kastrup J. Effect of mobilization of bone marrow stem cells by granulocyte colony stimulating factor on clinical symptoms, left ventricular perfusion and function in patients with severe chronic ischemic heart disease. Int J Cardiol. 2005;100:477– 483. Jørgensen E, Kelbæk H, Helqvist S, Jensen GV, Saunamaki K, Kastrup J, Havndrup O, Bundgaard H, Kyst MJ, Christiansen M, Andersen PS, Reiber JH. Predictors of coronary in-stent restenosis: importance of angiotensin-converting enzyme gene polymorphism and treatment with angiotensin-converting enzyme inhibitors. J Am Coll Cardiol. 2001;38: 1434 –1439. Kelbæk H, Thuesen L, Helqvist S, Kløvgaard L, Jørgensen E, Aljabbari S, Saunamaki K, Krussell LR, Jensen GVH, Bøtker HE, Lassen JF, Andersen HR, Thayssen P, Galløe A, van Weert AW. The Stenting Coronary Arteries in Non-Stress/Benestent Disease (SCANDSTENT) Trial. J Am Coll Cardiol. 2006;47:449 – 455. Jørgensen E, Ripa RS, Helqvist S, Wang Y, Johnsen HE, Grande P, Kastrup J. In-stent neo-intimal hyperplasia after stem cell mobilization by granulocyte-colony stimulating factor: preliminary intracoronary ultrasound results from a double-blind randomized placebo-controlled study of patients treated with percutaneous coronary intervention for ST-elevation myocardial infarction (STEMMI Trial). Int J Cardiol. July 28, 2005. DOI: 10.1016/j.ijcard.2005.06.045. Available at: http:// www.sciencedirect.com/science/journal/01675273. Accessed 2006. Kang HJ, Kim HS, Zhang SY, Park KW, Cho HJ, Koo BK, Kim YJ, Soo LD, Sohn DW, Han KS, Oh BH, Lee MM, Park YB. Effects of intracoronary infusion of peripheral blood stem-cells mobilised with granulocyte-colony stimulating factor on left ventricular systolic function and restenosis after coronary stenting in myocardial infarction: the MAGIC cell randomised clinical trial. Lancet. 2004;363:751–756.
CLINICAL PERSPECTIVE Previous research indicated that mobilization of bone marrow– derived stem cells with granulocyte-colony stimulating factor (G-CSF) could help regenerate cardiac cells after acute myocardial infarction by increased vascularization, regeneration of cardiomyocytes, and/or a direct antiapoptotic effect. Accordingly, we performed a randomized, placebo-controlled, double-blind trial to test whether patients with an ST-segment elevation myocardial infarction treated with primary percutaneous coronary stent intervention (PCI) derive therapeutic benefit from addition of G-CSF to modern conventional therapy. Disappointingly, there was no evidence of additional benefit of G-CSF after 6 months follow-up. Systolic wall thickening in the infarct area had increased 17% in the G-CSF group and 17% in the placebo group. Left ventricular ejection fraction increased at nearly identical rates in the 2 groups measured by both magnetic resonance imaging and echocardiography. Rates of death or myocardial infarction, new revascularization, and restenosis 6 months after the acute myocardial infarction were also similar in the G-CSF and placebo groups. Thus, subcutaneous G-CSF after PCI for ST-elevation infarction was safe, but did not lead to further improvements of left ventricular function when compared to placebo. It remains to be determined if G-CSF treatment could be an effective part of a treatment strategy combining several cytokines and/or local stem cell delivery. However, this trial underscores the need for blinding and placebo controls in the evaluation.
